[go: up one dir, main page]
Scribd
Upload
96 views1 page

Regression Analysis Homework Exercises

This document contains 4 questions from a homework assignment on regression analysis methods. Question 1 involves fitting a linear regression model to temperature and chemical yield data and performing various analyses. Question 2 involves fitting a linear model to other data and comparing steps from Question 1. Question 3 examines a subset of stock market data and assessing the linear fit. Question 4 involves fitting a weighted linear model to address non-constant error variance in other data.

Uploaded by

dm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
96 views1 page

Regression Analysis Homework Exercises

This document contains 4 questions from a homework assignment on regression analysis methods. Question 1 involves fitting a linear regression model to temperature and chemical yield data and performing various analyses. Question 2 involves fitting a linear model to other data and comparing steps from Question 1. Question 3 examines a subset of stock market data and assessing the linear fit. Question 4 involves fitting a weighted linear model to address non-constant error variance in other data.

Uploaded by

dm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Math 644 Regression Analysis Methods Homework 1

This collection of regression exercises is designed to reinforce understanding of the structure of the linear
model, the underlying assumptions, standard analysis, and interpretation of results. Some of the questions
require thinking about generalization of procedures discussed in class.

Question 1 A study was made on the effect of temperature on the yield of a chemical process. The following
data (coded) were collected:
𝑋: -5,-4,-3,-2,-1,0,1, 2, 3, 4, 5
𝑌 : 1, 5, 4, 7,10,8,9,13,14,13,18
1. Assuming a model, 𝑌 = 𝛽0 + 𝛽1 𝑋 + 𝜖, what are the least squares estimates of 𝛽0 and 𝛽1 ? What is the
prediction equation?
2. Construct the analysis of variance table and test the hypthesis, 𝐻0 : 𝛽1 = 0 at the 5% level.
3. What are the 95% confidence limits for 𝛽1 ?
4. What are the 95% confidence limts for 𝑌 when 𝑋 = 3?
5. What are the 95% confidence limits for the difference between the true mean value of 𝑌 when 𝑋 = 3 and
the true mean value when 𝑋 = −2? [Hint: Expand E[𝑌 ∣𝑋 = 3] − E[𝑌 ∣𝑋 = −2] and simplify.]
6. Are there any indications that a better model should be tried?
7. Comment on the number of levels of temperature investigated with respect to the estimate of 𝛽1 in the
assumed model.
[Draper and Smith, p55, Exercise A.]

Question 2 Using the matrix development, fit a straight line 𝑌 = 𝛽0 + 𝛽1 𝑋 + 𝜖 to the data below, and
provide a complete analysis. Use the steps in question 1 as a guide; what do you think of the applicability
of steps 1 and 7 here?
X: 1,1,2,3,4,4,4,5,6,6
Y: 4.2,3.8,3.0,2.3,1.8,2.0,2.2,2.0,2.5,2.7
[Draper and Smith, p138, Exercise E.]

Question 3 Comparing analysis of data subsets. The Amex - S&P500 data (Alexander p145 and spread-
sheet). Examine the subset of data corresponding to market moves in the range [-1%,1%]. How does the
relationship compare to the overall relationship? Include commentary on the reasonableness of a straight
line fit, homogeneity of variances, changes over time.

Question 4 Weighted Least Squares Problem.


Data:
𝑋: 1.15,1.90, 3,3,3,3,3, 5.34,5.38,5.4,5.4,5.45,
7.7,7.8,7.81,7.85,7.87,7.91,7.94, 9.03,9.07,9.11,9.14,9.16,9.37,
10.17,10.18,10.22,10.22,10.22,10.18,10.5,10.23,10.03,10.23
𝑌 : 0.99,0.98, 2.6,2.67,2.66,2.78,2.8, 5.92,5.35,4.33,4.89,5.21,
7.68,9.81,6.52,9.71,9.82,9.81,8.5, 9.47,11.45,12.14,11.5,10.65,10.64,
9.78,12.39,11.03,8,11.90,8.68,7.25,13.46,10.19,9.93
1. Fit the model 𝑦 = 𝛽0 + 𝛽1 𝑋 + 𝜖.
2. Plot the residuals from the model fit against fitted values, 𝑌ˆ , and predictor variable, 𝑋.
3. Suggest a weighting of the data that might control the apparent error variance inhomogeneity. Fit the
straight line model to the weighted data and examine the residuals.
[Draper and Smith, p112-115]

You might also like